Probing Defect Sites on TiO₂ with [Re₃(CO)₁₂H₃]: Spectroscopic Characterization of the Surface Species

Abstract

Samples of the anatase phase of titania were treated under vacuum to create Ti(3+) surface-defect sites and surface O(-) and O(2) (-) species (indicated by electron paramagnetic resonance (EPR) spectra), accompanied by the disappearance of bridging surface OH groups and the formation of terminal Ti(3+)-OH groups (indicated by IR spectra). EPR spectra showed that the probe molecule [Re(3)(CO)(12)H(3)] reacted preferentially with the Ti(3+) sites, forming Ti(4+) sites with OH groups as the [Re(3)(CO)(12)H(3)] was adsorbed. Extended X-ray absorption fine structure (EXAFS) spectra showed that these clusters were deprotonated upon adsorption, with the triangular metal frame remaining intact; EPR spectra demonstrated the simultaneous removal of surface O(-) and O(2) (-) species. The data determined by the three complementary techniques form the basis of a schematic representation of the surface chemistry. According to this picture, during evacuation at 773 K, defect sites are formed on hydroxylated titania as a bridging OH group is removed, forming two neighboring Ti(3+) sites, or, when a Ti(4+)-O bond is cleaved, forming a Ti(3+) site and an O(-) species, with the Ti(4+)-OH group being converted into a Ti(3+)-OH group. When the probe molecule [Re(3)(CO)(12)H(3)] is adsorbed on a titania surface with Ti(3+) defect sites, it reacts preferentially with these sites, becoming deprotonated, removing most of the oxygen radicals, and healing the defect sites.

Probing Surface Sites of TiO2: Reactions with [HRe(CO)5] and [CH3Re(CO)5]

“…Effect of titania treatment conditions on Ti 3 + site formation: consistency with the literature: In agreement with earlier reports, [9,18] our data show how the treatment conditions affect the titania surface-OH species. For example, when the titania was treated in flowing O 2 , the bands associated with OH groups on Ti 3 + sites were markedly reduced in intensity, relative to the same bands characterizing the titania that had been treated under vacuum (the reduction in band intensity is relative to the intensity of the band characteristic of the OH groups on Ti 4 + sites).…”

“…Smaller changes in the bands at ñ = 3672 and 3642 cm À1 were also observed. IR spectra characterizing the sample treated in flowing O 2 (TiO 2ox723 , Figure 1 B) indicate the presence of the same OH bands observed in the spectra of TiO 2vac723 ; however, consistent with previous observations, the intensity of the band at ñ = 3717 cm À1 relative to that of the Ti + 4 À OH bands was significantly reduced, [18,20] which is in line with the assignment of the band at ñ = 3717 cm À1 to Ti + 3 ÀOH groups that we would expect to have been almost completely removed by the treatment in O 2 . [9,22] In summary, the assignment of the band at ñ = 3717 cm À1 to Ti + 3 À OH is based on the intensity changes after treatments of the titania in O 2 or under vacuum and its position in the IR spectra.…”

“…Then it was treated at various temperatures under vacuum or in flowing O 2 , to provide samples containing various densities of surface Ti + 3 sites. [18] In the following, the titania samples are labeled according to their treatments. When the treatments took place in flowing O 2 , the samples are labeled with the subscript "ox" and when the treatment was under vacuum, the samples are labeled with the subscript "vac".…”

“…[36][37][38][39][40] Removal of Ti 12 ], [18] which presumably resulted 5 ]. This result is in agreement with the appearance of the band at ñ = 1623 cm À1 , which is also associated with water adsorbed on titania.…”

“…The literature reports were used as a basis for assigning these IR bands; the bands listed in the preceding sentence are assigned respectively to OH groups that are part of Ti 3 + À OH, Ti 4 + À OH, Ti 4 + À OH, and (Ti 4 + ) 2 À OH species and to OH groups in adsorbed water. [9,[18][19][20] Although bands at ñ = 3689 and 3672 cm À1 have both been assigned to Ti 4 + ÀOH groups, [9,19] it is clear from the spectra that each corresponds to separate modes, suggesting that the coordination of the Ti atoms to OH is different, corresponding to the different frequencies. [21] We lack sufficient information to determine the differences between these two sites.…”